Treatment strategies are stratified by risk factors;a goal of stratification into different prognostic groups is to design risk based treatment strategies. Long term survival is determined by "risk factors" that predict prognostic parameters such as probability of local control and risk of developing systemic or metastatic disease. Tumor hypoxia is predictive of both risk of metastases and aggressive local disease. A non-invasive assay to measure hypoxia would provide prognostic information regarding local tumor aggressiveness and metastatic risk for development of risk based therapy, and for monitoring changes in oxygenation with treatment, could impact further therapy. The central hypothesis of this proposal is that tumor hypoxia and changes in oxygenation can be evaluated non-invasively using selected surrogate "markers". In all studies, we will utilize a stereotaxic template we have developed to register both the in vivo data and the p02 and pimonidazole studies. In Aim 1 we will test and validate a novel derivative of misonidazole (trifluoromisonidazole (T19F-FMISO)) for imaging hypoxia. We will determine the optimal dose as a balance between signal to noise requirements vs. specificity for imaging hypoxia, and also compare to 18F-misonidazole and p02. In Aim 2 we will evaluate quantitation of lactate, dynamic contrast enhanced MRI, andT19F-FMISO as oxygen surrogates and validate them against p02 and pimonidazole. In Aim 2B, we will study changes in hypoxia induced by anti-neoplastic therapy and use these measurements as potential surrogate and compare to p02, microvessel density and radiobiological assays. The goal of Aim 2 is to determine which is the best surrogate of hypoxia and apply this in Aim 3. Aim 3 will use this data to optimize hypoxia driven suicide gene therapy. In aim 3, we will develop a fusion suicide gene (Cytosine Deaminase - Uracil Phosphoribosyl Transferase - CD-UPRT), under the control of a hypoxia response element (HRE) which can be quantitatively imaged by 19F NMR chemical shift imaging. We will develop this system as a both a reporter and suicide therapy system and evaluate its efficacy and compare it with thymidine kinase in parallel studies.